paper-with-me

Papers

ConTNet: Why not use convolution and transformer at the same time?

2021-04-27 · Haotian Yan, Zhe Li, Weijian Li, Changhu Wang, Ming Wu, Chuang Zhang

Although convolutional networks (ConvNets) have enjoyed great success in computer vision (CV), it suffers from capturing global information crucial to dense prediction tasks such as object detection and segmentation. In this work, we innovatively propose ConTNet (ConvolutionTransformer Network), combining transformer with ConvNet architectures to provide large receptive fields. Unlike the recently-proposed transformer-based models (e.g., ViT, DeiT) that are sensitive to hyper-parameters and extremely dependent on a pile of data augmentations when trained from scratch on a midsize dataset (e.g., ImageNet1k), ConTNet can be optimized like normal ConvNets (e.g., ResNet) and preserve an outstanding robustness. It is also worth pointing that, given identical strong data augmentations, the performance improvement of ConTNet is more remarkable than that of ResNet. We present its superiority and effectiveness on image classification and downstream tasks. For example, our ConTNet achieves 81.8% top-1 accuracy on ImageNet which is the same as DeiT-B with less than 40% computational complexity. ConTNet-M also outperforms ResNet50 as the backbone of both Faster-RCNN (by 2.6%) and Mask-RCNN (by 3.2%) on COCO2017 dataset. We hope that ConTNet could serve as a useful backbone for CV tasks and bring new ideas for model design

📄 PDF Abstract BibTeX arXiv:2104.13497

Code (2)

yan-hao-tian/ConTNet 공식 구현 pytorch
mindspore-courses/External-Attention-MindSpore/blob/main/model/backbone/ConTNet.py mindspore

Tasks

image-classificationImage Classificationobject-detectionObject Detection

Methods 이 논문이 사용한 방법론

ReLU How Do I Communicate to Expedia? How Do I Communicate to Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Live Support & Special Travel…
Residual Connection 설명 없음
Max Pooling Max Pooling is a pooling operation that calculates the maximum value for patches of a feature map, and uses it to create a downsampled (pooled) feature map. It is usually…
Convolution A convolution is a type of matrix operation, consisting of a kernel, a small matrix of weights, that slides over input data performing element-wise multiplication with the…
1x1 Convolution A 1 x 1 Convolution is a convolution with some special properties in that it can be used for dimensionality reduction,…
Average Pooling 설명 없음
Residual Block Residual Blocks are skip-connection blocks that learn residual functions with reference to the layer inputs, instead of learning unreferenced functions. They were introduced…
Batch Normalization 설명 없음

Similar Papers 제목 키워드 기반

Wake Word Detection with Streaming Transformers

2021-02-08 · Yiming Wang, Hang Lv, Daniel Povey, Lei Xie 외

Modern wake word detection systems usually rely on neural networks for acoustic modeling. Transformers has recently shown superior performance over LSTM and convolutional networks in various sequence modeling tasks with …

Co-advise: Cross Inductive Bias Distillation

2021-06-23 · CVPR 2022 1 · Sucheng Ren, Zhengqi Gao, Tianyu Hua, Zihui Xue 외

Transformers recently are adapted from the community of natural language processing as a promising substitute of convolution-based neural networks for visual learning tasks. However, its supremacy degenerates given an in…

Inductive Bias

Combining Transformer Generators with Convolutional Discriminators

2021-05-21 · Ricard Durall, Stanislav Frolov, Jörn Hees, Federico Raue 외

Transformer models have recently attracted much interest from computer vision researchers and have since been successfully employed for several problems traditionally addressed with convolutional neural networks. At the …

Data AugmentationImage GenerationSuper-Resolution

MaskConver: Revisiting Pure Convolution Model for Panoptic Segmentation

2023-12-11 · Abdullah Rashwan, Jiageng Zhang, Ali Taalimi, Fan Yang 외

In recent years, transformer-based models have dominated panoptic segmentation, thanks to their strong modeling capabilities and their unified representation for both semantic and instance classes as global binary masks.…

DecodermodelPanoptic Segmentation

PU-EdgeFormer: Edge Transformer for Dense Prediction in Point Cloud Upsampling

2023-05-02 · Dohoon Kim, Minwoo Shin, Joonki Paik

Despite the recent development of deep learning-based point cloud upsampling, most MLP-based point cloud upsampling methods have limitations in that it is difficult to train the local and global structure of the point cl…

point cloud upsampling